Page 13 - Computational Statistics Handbook with MATLAB

Preface

Computational statistics is a fascinating and relatively new field within sta-
tistics. While much of classical statistics relies on parameterized functions
and related assumptions, the computational statistics approach is to let the
data tell the story. The advent of computers with their number-crunching
capability, as well as their power to show on the screen two- and three-
dimensional structures, has made computational statistics available for any
data analyst to use.
Computational statistics has a lot to offer the researcher faced with a file
full of numbers. The methods of computational statistics can provide assis-
tance ranging from preliminary exploratory data analysis to sophisticated
probability density estimation techniques, Monte Carlo methods, and pow-
erful multi-dimensional visualization. All of this power and novel ways of
looking at data are accessible to researchers in their daily data analysis tasks.
One purpose of this book is to facilitate the exploration of these methods and
approaches and to provide the tools to make of this, not just a theoretical
exploration, but a practical one. The two main goals of this book are:

• To make computational statistics techniques available to a wide
range of users, including engineers and scientists, and
• To promote the use of MATLAB® by statisticians and other data
analysts.

MATLAB and Handle Graphics® are registered trademarks of
The MathWorks, Inc.
There are wonderful books that cover many of the techniques in computa-
tional statistics and, in the course of this book, references will be made to
many of them. However, there are very few books that have endeavored to
forgo the theoretical underpinnings to present the methods and techniques in
a manner immediately usable to the practitioner. The approach we take in
this book is to make computational statistics accessible to a wide range of
users and to provide an understanding of statistics from a computational
point of view via algorithms applied to real applications.
This book is intended for researchers in engineering, statistics, psychology,
biostatistics, data mining and any other discipline that must deal with the
analysis of raw data. Students at the senior undergraduate level or beginning
graduate level in statistics or engineering can use the book to supplement
course material. Exercises are included with each chapter, making it suitable
as a textbook for a course in computational statistics and data analysis. Scien-

© 2002 by Chapman & Hall/CRC